Meta describes engagement bait as spammy posts that goad people into interacting with likes, shares, comments and other actions. [1] That is a company's own language for behaviour on its own service, not a definition handed down by a dictionary or a researcher, and it is worth being plain about that before anything else on this page: this is Meta describing what it demotes, not a neutral account of what the tactic is.
It earns its place in the bait family for the reason the family is named for. Nothing about it is really addressed to you as a person with something to say. It exists to be interacted with, and the interaction is what it was built to collect — not evidence that you found it funny, agreed with it, or thought it was worth a comment.
What Meta says it is for
Meta's own account of the mechanism is this: it says the tactic seeks to take advantage of its News Feed algorithm by boosting engagement in order to get greater reach. [2] That is Meta's stated characterisation of intent — its own explanation for why someone would do this — and it is not the same thing as a measured finding about what any particular post accomplishes. Treat it as what a company believes motivates the pattern it is policing, not as a fact established about every post that resembles one.
A 2017 announcement, not a live status report
In a December 2017 announcement, Meta said it would begin demoting individual posts that use the tactic, and that posts using it would be shown less in News Feed, with stricter demotions rolled out over following weeks for pages that used it systematically and repeatedly. [3]
That is a record of what a company said it would build, in a specific year, not a current account of how anything is ranked today. Nothing here is evidence that the demotion described still runs, that it ever worked as described, or that a post you are looking at right now is being suppressed because of it. Ranking systems change constantly and mostly without public notice; an announcement about a machine-learning classifier from 2017 tells you what one company intended to ship that year, and nothing reliable about News Feed today.
Recognising it
- A request for one specific action — a like, a share, a tag, a particular kind of comment — with little else being said
- The action costs nothing to fake an answer to: "comment your birth month" needs no truth, no thought and no risk from whoever supplies it
- The same shape recurs across many posts from the same account or page, swapping only the prompt
- Whether the account keeps posting this way depends on nothing you do — someone else supplies the numbers if you do not
Every one of these has an innocent reading. People run genuine polls, genuinely want to know your star sign, and genuinely enjoy participatory nonsense with no algorithm in mind at all. A single post asking for a like tells you almost nothing on its own; a pattern, repeated at volume with nothing else behind it, is closer to what the label is actually describing.
The honest difficulty
Plenty of completely sincere posts ask for engagement outright, and asking is not, by itself, evidence of anything. A birthday post inviting comments, a creator asking followers to share something they made, a genuine "tag a friend who needs to see this" — all of it asks for the same actions this page is naming, from people who mean every word of it.
What separates that from the tactic is the engineering: a post built and repeated specifically to produce the metric, rather than one that happens to ask for an interaction along the way. That distinction is close to impossible to make from a single post, and it is ours to draw, not Meta's — its own wording describes the spammy version and says nothing about how to tell it apart from the sincere one, which is the harder and far more common judgement call a reader is actually left with.
What to do
Don't supply the specific action. The tactic wants a like, a share, a tag or a comment — not you particularly, but the count going up. Not doing what it is asking denies it the one input it needs. No one has measured whether declining actually changes how often an account keeps posting this way; it is an argument from what the request is for, not a tested result.
Don't treat a platform's word as today's behaviour. Whatever a company announced about demoting posts like this, nothing here tells you it is currently enforced, so "the algorithm will handle it" is not a reason to either engage or relax.
Notice the pattern, not the post. One request for a like tells you very little on its own. The same account running the same request format repeatedly, with nothing else to say, is the more genuine signal — and even then, declining to supply the interaction costs you nothing either way.